It is well established that prokaryotes regulate the transcription of many of their genes in response to changes in the organisms' environment (J. B. Stock, A. M. Stock, and J. M. Mottonen, Nature, 344, 395-400 (1990)). Such regulation is essential if the organism is to adapt itself to survival in a changing environment, and pathogenic bacteria rely on such regulatory systems to enable them to survive within their host's body (J. F. Miller, J. J. Mekalanos, S. Falkow, Science, 243, 1059 (1989). Chemical compounds that interfere with the regulatory mechanisms would be expected to be useful anti-infective drugs, as they would prevent bacteria from making necessary adaptive changes in their patterns of gene expression.
Virulence, chemotaxis, toxin production, sporulation, and reproduction are examples of the bacterial processes that are under regulatory control, and which could be inhibited by such compounds. The inhibition of one or more of these processes is expected to lead to reduced virulence, a slowing or halting of bacterial growth and reproduction, and even to bacterial cell death if vital functions are interrupted.
For example, it has been shown that Salmonella species express certain proteins, under regulatory control and in response to the presence of epithelial cells, which enable them to adhere to and invade intestinal epithelial cells. Bacteria that are unable to synthesize these proteins are avirulent: they cannot cause infection in mice (B. B. Finlay, F. Heffron, S. Falkow, Science, 243, 940-943 (1989)). A similar effect would be expected if the genes coding for these proteins were intact, but remained unexpressed.
To accomplish adaptive responses to the environment, bacteria rely on phosphorelay mechanisms, referred to in the art as a "two-component switches." These switches have the net effect of transmitting information from the environment to the cell nucleus, where the information is responded to by the switching on or off of transcription of relevant genes. The first step of this phosphorelay scheme relies on numerous histidine protein kinase (HPK) enzymes. Most of these HPK enzymes are sensor molecules, and respond to stimulation by specific environmental signals by transferring phosphate from ATP to a histidine residue of the HPK protein. Some HPK enzymes are stimulated by the presence of acceptor proteins (described below), the concentration of which are modulated by environmental signals. In either case, this auto-phosphorylation is followed by transfer of the phosphate to an aspartyl residue of one or more acceptor proteins (the second components of the two-component switch), which are either regulators of gene expression (by binding to control regions on DNA, or to the RNA polymerase complex) or are themselves kinases for other acceptor molecules. These secondary acceptors may again be regulatory proteins, or kinases toward yet another protein. This cascade of phosphate from protein to protein eventually results in the phosphorylation of one or more regulatory proteins, which then control gene expression.
Mammalian cells do not, or at least are not presently known to, utilize HPK-driven phosphorelay systems for gene regulation. Thus, compounds which selectively inhibit either the autophosphorylation of the HPK protein, or the phosphotransfer step(s), or both, would not be expected to have undesirable effects on the host organism, and are promising candidates for antiinfective drugs. The emergence of drug-resistant pathogenic organisms that are resistant to one or more of the currently available drugs has created a need for novel antibiotics, that act by mechanisms unrelated to those of currently available agents, and inhibitors of HPK would fill this need. The presence of multiple HPK-driven systems (over fifty are currently known) in bacteria gives HPK inhibitors a potential advantage over current antibiotics, in that mutations of a single HPK enzyme are unlikely to confer drug resistance to an organism.
Recently, workers in this field reported a method for detecting bacterial "virulence" genes that are selectively expressed when bacteria infect a host (M. J. Mahan, J. M. Slauch, and J. J. Mekalanos, Science, 259, 686-688 (1993)). The potential use of this information in the design of new antibiotics was mentioned, but actual methods of reducing expression of these genes were not described. A preliminary report from another group of workers disclosed inhibitors of the two-component switch controlling alginate gene activation in Pseudomonas aeruginosa in an in vitro system (S. Roychoudhury et al., Proc. Nat. Acad. Sci., 90, 965-969 (1993)), but no anti-bacterial activity of the compounds was reported.